Technological Improvements in the management of type 1 diabetes (T1D), including insulin analogs, insulin pump therapy, and most recently, continuous glucose monitoring, are still not sufficient to normalize glucose levels in most people with T1D. It is likely that no manual, "open-loop" therapeutic insulin regimen will be able to optimally control glycemia in patients with T1D, and biological islet-cell replacement is presently unfeasible. "Closed-loop" automated artificial pancreas systems, consisting of an insulin pump to precisely deliver variable amounts of insulin, a continuous glucose sensor to accurately determine the glucose levels, and effective algorithms to determine insulin delivery rates based on real-time glucose readings, remains the most promising intervention to reduce hyperglycemic and hypoglycemic exposures. Short-duration hospital-based studies in adults and adolescents have demonstrated feasibility of this approach. However, deficiencies in current closed-loop approaches have become apparent. Present methods of subcutaneous insulin delivery are not sufficiently rapid to prevent meal-related increases in blood glucose levels, and persistent elevations in plasma insulin levels are associated with a risk of subsequent hypoglycemia. Most importantly, inpatient studies cannot realistically replicate the variability of meals and activities inherent in the usual home environment. The objectives of this research project are twofold: (1) to optimize the performance of closed-loop systems using strategies to either accelerate subcutaneous insulin delivery or delay carbohydrate absorption, thereby allowing improved prandial glucose control and potentially limiting hypoglycemia;and (2) to evaluate the safety and effectiveness of closed-loop delivery in the ambulatory environment, using a stepwise approach that gradually increases utilization of remote computerized and wireless communication techniques and decreases reliance on traditional methods of subject monitoring. The planned experiments will build on our expertise in insulin pharmacokinetics and pharmacodynamics and glucose counter-regulation and benefit from our ongoing industry collaborations. The proposed studies are novel, important, and likely to advance our understanding not only of closed-loop control of T1D, but also its potential use as a treatment for hypoglycemia unawareness. PUBLIC HEALTH RELEVANCE: For the approximately 1 in 700 Americans with type 1 diabetes, recommended intensive treatments are difficult, burdensome, and frequently insufficient to prevent long-term complications and morbidity. Development of automated feedback, glucose-controlled, insulin delivery systems that would effectively regulate blood sugar levels would lead to improvements in patient outcomes and quality of life, and potentially, reduction in societal health care costs.